sum.py

"""
Given a list of numbers, compute their sum.
"""
import pandas as pd
from numba import jit
import numpy as np

from typing import List

from utils.profiler import time_this, timed_report
from utils.profiler import ExponentialRange

def random_numeric_list(n: int) -> List[np.float64]:
    return list(np.random.random(n))


@time_this(lambda x: len(x))
def fast_sum(values: List[float]) -> float:
    accum = 0
    for value in values:
        accum += value
    return accum

@time_this(lambda x: len(x))
def fast_native_sum(values: List[float]) -> float:
    return sum(values)

@time_this(lambda x: len(x))
def numpy_fast_sum(values: np.ndarray) -> float:
    return np.sum(values)

@time_this(lambda x: len(x))
def pandas_fast_sum(values: pd.Series) -> float:
    return values.sum()

@jit(nopython=True)
def _numba_fast_sum(values: np.ndarray) -> float:
    """
    JIT compilation occurs after one use
    """
    accum = 0
    for value in values:
        accum += value
    return accum

# Force numba to run jit compilation
_numba_fast_sum(np.array(random_numeric_list(100000)))


@time_this(lambda x: len(x))
def numba_fast_sum(values: np.ndarray) -> float:
    """
    Redeclare optimized wrapper
    """
    return _numba_fast_sum(values)

if __name__ == '__main__':

    exp_range = ExponentialRange(0, 8, 1/4)
    values = random_numeric_list(exp_range.max)

    with timed_report():
        for i in exp_range.iterator():
            fast_sum(values[:i])

        for i in exp_range.iterator():
            fast_native_sum(values[:i])

        for i in exp_range.iterator():
            numpy_fast_sum(np.array(values[:i]))

        for i in exp_range.iterator():
            pandas_fast_sum(pd.Series(values[:i]))

        for i in exp_range.iterator():
            numba_fast_sum(np.array(values[:i]))